發(fā)布時間：2021-03-07 05:12

　　隨著信息技術(shù)的不斷發(fā)展和基礎(chǔ)設(shè)施的不斷完善,大數(shù)據(jù)技術(shù)已廣泛應(yīng)用于各個行業(yè),比如醫(yī)療、教育、餐飲、物流、汽車、金融和娛樂等行業(yè),給人們的生活帶來諸多便利。在大學(xué),隨著管理手段信息化的不斷深入,產(chǎn)生了大量的數(shù)據(jù),其中,大學(xué)生日常生活和學(xué)習(xí)行為所累積的數(shù)據(jù)引起了高校管理人員的高度重視,也成為廣大研究者的研究對象。通過對這些數(shù)據(jù)進行處理和分析,則可以獲得學(xué)生的行為特征和規(guī)律,為學(xué)生管理者更好地管理學(xué)生提供參考。本文基于蘭州理工大學(xué)的學(xué)生數(shù)據(jù),其中包含學(xué)生的書籍借閱、校園卡消費、兩學(xué)年的成績和學(xué)生專業(yè)記錄等數(shù)據(jù),作為要調(diào)查的數(shù)據(jù)源。使用RapidMiner數(shù)據(jù)架構(gòu)框架,可以預(yù)處理數(shù)據(jù)集并集成不同的數(shù)據(jù)源以獲得一組數(shù)據(jù)以進行分析。進行的主要工作如下:（1）利用FP—growth算法來挖掘?qū)W生的學(xué)習(xí)成績、借閱的書籍數(shù)量、不同專業(yè)與校園卡消費之間的相關(guān)關(guān)系,來預(yù)測學(xué)生行為。還使用Python Pandas軟件包進行了統(tǒng)計分析,以確保數(shù)據(jù)平衡以及檢測和處理任何異常值。（2）通過使用K-means算法對學(xué)生數(shù)據(jù)進行聚類,根據(jù)聚類結(jié)果挖掘不同學(xué)生的學(xué)業(yè)成績、圖書借閱數(shù)據(jù)與校園卡消費數(shù)據(jù)之間的關(guān)系,以及不同... 

【文章來源】：蘭州理工大學(xué)甘肅省

【文章頁數(shù)】：76 頁

【學(xué)位級別】：碩士

【文章目錄】：
Abstract
摘要
Chapter1 Introduction
    1.1 Research Background and Significance
        1.1.1 Research Background
        1.1.2 Research Significance
    1.2 Research status
        1.2.1 Research Status of Big Data
        1.2.2 Research Status of Association Rules
        1.2.3 Research Status of Cluster Analysis
    1.3 Research content
    1.4 Thesis Structure
Chapter2 Related Theory and Technology Review
    2.1 Data Mining
    2.2 Data Preprocessing
    2.3 Association Rules Mining
    2.4 Cluster Analysis
    2.5 Classification Techniques
        2.5.1 Neural Networks Model
        2.5.2 Na?ve Bayes Model
        2.5.3 Support Vector Machine Model
        2.5.4 Random Forest Model
        2.5.5 K-Fold Cross Validation Technique
    2.6 Data Mining tool
    2.7 Chapter Summary
Chapter3 Preprocessing of Student Behavior Data
    3.1 Introduction
    3.2 Data introduction
    3.3 Data Cleaning
        3.3.1 Library Book borrowing data cleaning
        3.3.2 Card Consumption Data Cleaning
        3.3.3 Grade Data Cleaning
        3.3.4 University Departments Data Cleaning
    3.4 Data Integration
    3.5 Chapter Summary
Chapter4 Study on the relevance of student behavior
    4.1 Introduction
    4.2 Association rules of student behavior
        4.2.1 FP-Growth Algorithm
        4.2.2 Discretization of students’Behavior data
        4.2.3 Relevance analysis of behavior data
    4.3 Cluster Analysis of student behavior
        4.3.1 K-means algorithm
        4.3.2 Cluster analysis based on K-means algorithm
    4.4 Chapter summary
Chapter5 Predicting Student Performance
    5.1 Introduction
    5.2 Classification and prediction models
        5.2.1 Neural Networks Model
        5.2.2 Na?ve Bayes Model
        5.2.3 Support Vector Machine Model
        5.2.4 Random Forest Model
        5.2.5 K-Fold Cross Validation
    5.3 Prediction Models Evaluation Matrices of students’data
    5.4 Predictive analysis of student data
        5.4.1 Proposed model for predicting student performance
        5.4.2 Proposed model Evaluation:
        5.4.3 Student performance predictive models’comparison
        5.4.4 Student performance predictive models’comparison for students’majors
    5.5 Chapter summary
Conclusion and Future Work
References
Acknowledgement
Academic papers and awards



本文編號：3068459